Paroxysmal nocturnal hemoglobinuria (PNH) arises as the result of a somatic mutation affecting a pluripotent hematopoietic stem cell. The affected gene, called PIG-A (phosphatidylinositol glycan-class A) encodes a protein essential for normal biosynthesis of the glycosyl phosphatidylinositol (GPI) moiety that serves as a membrane anchor for many cellular proteins. The PIG-A mutations cause GPI-anchored proteins (GPI-AP) to be deficient on the hematopoietic elements of PNH. The deficiency of GPI-AP regulation of complement accounts for the hemolysis that is the hallmark of PNH. PNH is a clonal disease, and the circulating hematopoietic elements are a mosaic of normal and abnormal cells. For clinical evidence to become apparent, the PNH clone must expand so that signs and symptoms of hemolysis are observed. Thus, the abnormal clone appears to have a proliferative advantage. In Specific Aim I we propose to determine if a mutation of PIG-A provides a proliferative advantage in vivo. Homologous recombination using two step gene replacement will be used to inactivate the gene after the mouse has fully developed. In Specific Aim II, evidence for expression of PNH specific genes will be sought. Lymphocyte clones from a patient with PNH have been developed. Some clones normally express of GPI-AP (putatively the progeny of residual normal stem cells) and others are deficient (progeny of the abnormal clone). These clones should be genetically matched except for PNH specific differences. PNH specific genes will be identified and characterized using differential display polymerase chain reaction. The only curative therapy for PNH is marrow ablative chemotherapy followed by allogeneic bone marrow transplantation. In essentially all patients with PNH,. cells with normal expression of GPI-AP circulate along with cells that are GPI-AP deficient, suggesting that residual normal stem cells are present. If normal stem cells can be separated from abnormal stem cells, they could be used in autologous bone marrow transplantation. In Specific Aim II, we will determine if PNH stem cells can be distinguished from normal. Studies aimed at determining the feasibility of purging GPI-AP negative cells from a mixture of GPI-AP positive and CPI-AP negative cells are also described.